completed
Open ledger for agricultural land
Current Project Status
Complete
Amount
Received
$55,000
Amount
Requested
$55,000
Percentage
Received
100.00%
Solution

Automatic detection of field boundaries and seeded acres using super-high resolution SatEO initially in Tanzania open to smallholder farmers

Problem

Current problem is outdated & inaccurate agricultural field boundaries created, managed and updated by national Cadastral agencies manually.

Addresses Challenge
Feasibility
Auditability

Team

1 member

Open ledger for agricultural land

DigiFarm's solution will enable highly accurate and unbiased classification of agricultural land using up-to-date super-high resolution Satellite data (1m per pixel) in order to automatically detect field boundaries and seeded acres.

The solution will address the "Property Registration" section of the challenge. Furthermore, the solution will enable smallholder farmers to easier access financing, agronomic advisory and build their credit profile. Furthermore, with nearly 80% of households in Tanzania engaging in agriculture and at least one third gaining more than half of their income from agricultural activities, access to finance for small-scale producers is a major catalyst to broad based economic growth.

For a long time and especially in traditional forms of financing, one of the key limiting factors for access to loans for smallholder farmers has been lack of collateral. Looking at land ownership registration for instance, data collected by Tanzania’s bureau of statistics in 2018 shows that out of 8.7 million farms surveyed only 18% were registered.

The current problem is the lack of historical and in-season data to assess credit risk on smallholder farmers, this due to a lack of infrastructure consisting of agricultural land classification, crop classification, long-term productivity assessment (20-30+ years) on the individual farm-land and field boundaries.

Currently, field boundaries are manually created by field agents walking the corners of a physical agricultural field and geo-tagging those boundaries, this is time-consuming, expensive and often inaccurate. In order to provide a reliable and affordable solution the only way is to use automate this through the use of deep learning object detection and high-resolution Satellite data.

DigiFarm's solution will enable and empower smallholder farmers through open access to:

  1. Detecting in-season field boundaries and seeded acres - will help farmers assess and plan their growing season accurately through accessing the crop-inputs required (seeds, fertiliser, crop protection).
  2. Long-term productivity assessment - DigiFarm will provide 35 years of long-term productivity data (NDVI) for each individual field, including slope (growth of plant biomass over that period) which can be used to access financing and build a credit rating.
  3. In-season detection of management zones - DigiFarm will provide in-field assessment of the low, medium and high productivity zones (EVI/biomass) in order for the farmer to assess the problem areas and ability to apply variable fertiliser/fungicide application.

The main risks associated with this project is:

Type of risk: Technological

Description of the risk: Inability for deep neural network models to not accurately detect and predict field boundaries and seeded acres at a high accuracy in smallholder markets such as Tanzania where field boundaries are typically small and dynamic, due to extensive agricultural practices differences and variability, topography, crops, agricultural practice and weather (cloud-cover) restricting the model to interpret the imagery-data accurately, and accuracy resulting in lower than 0.96 (Intersection over Union) for field boundary detection.

Effect of the risk? Effect of risk includes that the field boundary detection model does not reach high enough accuracy in order to be commercially attractive to clients, i.e. not high enough accuracy compared to Cadastral map data or manual delineation.

Mitigation methods? Solution; develop regional models which provide the highest level of accuracy for key markets, to collect a substantial amount of training data (manual delineation) for the targeted areas. Additionally, DigiFarm super-resolves Sentinel-2 from 10m to 1m per pixel resolution which will be critical in order to properly delineate the field boundaries.

Type of risk: Cost

Description of risk: computational complexity and costs of data processing too high for planned price-point and practical market applications, limiting uptake and market segment.

Effect of the risk: if data processing costs are too high which will reduce our profit margins, this will risk the long-term sustainability of DigiFarm’s business model and operational capacity, as DigiFarm has identified the price-point with extensive market research to determine the (lowest possible price), currently 25% lower than competitors for 10x higher resolution and accuracy.

Mitigation methods: Data engineering and algorithmic design is continuously focusing on computational efficiency and scalability. DigiFarm will set up its own HPC infrastructure with local GPU-instances in order to reduce costs of data processing in cloud providers ecosystems where costs are ~$3 per hour per GPU, compared to $0.30 per hour. This has only been made possible due to: (a) recent advances in hardware (Graphic Processing Units) by industry (lead by NVIDIA) where performance of GPUs have tripled since 2015, (200 TFLOPS to 1500) combined with the affordability of the units (approx. 300% reduction in performance/$ ratio).

<u>Roadmap:</u>

<u>M1-M2</u>

  1. Identify AOI in Tanzania in conjunction with UNCDF local team where ground truth data is available (from small section of small holder farmers)
  2. Process Sentinel-2 imagery for 2022 data (3-4 cloud free images, L2A) across the AOI
  3. Deeply resolve the Sentinel-2 10m per pixel imagery to 1m per pixel (4-bands, RGB + NIR)
  4. Start collecting training data (manual delineation of field boundaries across the AOI)
  5. Start model training (deep neural network) based on the training data

<u>M2-M4</u>

  1. Proof-of-concept: field boundaries and seeded acres first model results
  2. Re-train model if necessary in order to achieve accuracy of IoU of 0.95+
  3. Technical architecture documentation posted at https://digifarming.readme.io/
  4. Automatically detect long-term productivity Zones (35 years) on the AOI
  5. Coordinate with local financial institutions and credit bureau's to assess credit portfolio of smallholder farmers - in order to build a model where farmers can access affordable and reliable financing pre-season and during-season
  6. Proof-of-concept: mobile DigiFarm
  7. Formal partnership with legal professional(s) to liaise with Tanzania land management office
  8. Blog post: write-up
  9. Blog post: pilot scenario:

<u>M4-M6</u>

  1. Extend AOI to all agricultural land in Tanzania
  2. Open the database to smallholder farmers to access this data and to communicate data to the financial and credit lending facilities
  3. Blog post: "DigiFarm in Tanzania - Results and findings"
  4. DigiFarm Whitepaper (overview of solution concept and technical architecture)

End of Fund 8 grant

<u>M6-M12</u>

  1. DigiFarm Whitepaper v2: Decentralized registry of field boundary productivity in Tanzania
  2. DigiFarm Initial Stake Pool Offering (ISPO)
  3. DigiFarm beta Dapp on Testnet
  4. 150 Tanzania property owners field testing DigiFarm
  5. Secure partnerships with local financial institutions and credit lending facilities

<u>M12+</u>

  1. DigiFarm Level 3 Certified Dapp
  2. DigiFarm launch on Cardano mainnet
  3. 1,000 registered smallholder farmers in Tanzania
  4. DigiFarm's platform fully operational
  5. Expand to additional regions incl. Kenya and Nigeria

Budget breakdown:

Hosting for the project website and code repositories are provided free of charge via Github. Community outreach will be done via (free) Linkedin, Facebook and YouTube accounts along with Project Catalyst communication channels.

Detailed roadmap above for descriptions of the tasks and work products that will be delivered in three, four-week sprints

  • Month 1-2 (Tasks #1-5): 150 hours x $65 = $9,750
  • Month 2-4 (Tasks #1-9): 350 hours x $65 = $22,750
  • Month 4-6 (Tasks #1-4): 200 hours x $65 = $13,000
  • Professional fees: including legal, accounting, and communications: $8,000
  • SUBTOTAL: $53,500
  • plus 10% project management and office expenses: $5,350

TOTAL: <u>$58,850</u>

DigiFarm’s team is the ideal fit for the project as our core team has extensive experience in (a) developing agricultural technology for crop-monitoring using AI and remote sensing (Satellite data) to the agribusinesses market (B2B/B2G) using SaaS-models. Successfully built commercial agricultural technological solutions using remote sensing (Satellite-data) and AI across 100 million hectares: >90% accuracy in crop Detection and >85% accuracy in yield-prediction in soybean and corn (US/Brazil) (b) core team has over 15+ years of on-the-ground crop-producing (farming) experience and close partnership withs Felleskjøpet (largest ag-coop in Norway, NLR (Norwegian Agricultural Advisory Organisation) and University of Life Sciences (NMBU) (c) commercial and corporate Ag-market: over 20+ years combined corporate agriculture leadership experience (d) over 40+ experience in agronomy academic research internationally.

Additional qualifications in DigiFarm’s core team and founders (10) include technical and agronomical experience: (a) over 40 years combined international work experience in precision-Ag projects in Canada, USA, Germany, Switzerland, Brazil, Australia, Russia and Ukraine (b) successfully filed 5 patents (AI-based technologies) in agriculture/biology (e) developed technology for Zoner.ag (one of first geospatial web-platforms for analyzing agricultural fields) successfully acquired by Bayer to become the geospatial engine of Xarvio digital-farming platform (owned by BASF). Management capacity: led and managed the Bayer CropScience division as Global Technology Lead with the Digital Farming Division, overseeing expansion Xarvio to over 100 employees, serving over 3.4 million farmers and agronomists worldwide (b) founded and grew AI-based Gamaya (Swiss-based) agtech startup, managed team growth to 45 employees in under 24 months and secured $20 million in VC funding from Mahindra.

  • Nils Helset, 15th generation farmer in Norway. Extensive experience with precision agriculture services and agronomy - over 8 years experience in crop-producing, managing over 40k decares of pilot farmers with Felleskjøpet.
  • Konstantin Varik - built advanced AI models in agriculture using remote sensing for over 8 years including AI model for crop-yield prediction for maize and soybeans for USA, Brazil and Argentina with 98-99% yield-prediction accuracy 1-1.5 months prior to harvesting (data provided to the USDA).
  • Alex Melnichouck founded B2B ag-tech startups Zoner.ag in 2012 (acquired by BASF in 2015), led and managed the global BASF digital farming team for over 4 years.
  • Yosef Akhtman founded and grew Gamaya, Swiss-based ag-tech startup focused on remote sensing and AI funded by Mahindra. Yosef is a scientist and inventor with over 5 registered patents in agriculture and biology.

Progress of the development of the project will be measured by:

  1. Github commits for each sprint (4 weeks)
  2. Reports and deliverables at each milestone (M1-M6)
  3. Prototype and web-application
  4. Onboarded users and feedback looped

Progress of the development of the project will be measured by:

  1. Ability to reach target IoU accuracy of delineation of permanent crops across all arable land in Tanzania at above IoU of 0.95
  2. Ability to accurately assess and calibrate with local farmers and ground truth data, e.g. yield-data and productivity zones, e.g. low/medium and high zones will be assigned predictive yield-values, in-season and historically above 90% accuracy which can be used as leverage from securing financing for the smallholder farmers.
  3. Ability to secure 500+ farmers within first 9 months and secure 5 lending institutions as pilot users
  4. Successful sustainable commercial model after the grant-period has ended
  5. Further geographical coverage including Nigeria and Kenya

No, this is an entirely new proposal.

close

Playlist

  • EP2: epoch_length

    Authored by: Darlington Kofa

    3m 24s
    Darlington Kofa
  • EP1: 'd' parameter

    Authored by: Darlington Kofa

    4m 3s
    Darlington Kofa
  • EP3: key_deposit

    Authored by: Darlington Kofa

    3m 48s
    Darlington Kofa
  • EP4: epoch_no

    Authored by: Darlington Kofa

    2m 16s
    Darlington Kofa
  • EP5: max_block_size

    Authored by: Darlington Kofa

    3m 14s
    Darlington Kofa
  • EP6: pool_deposit

    Authored by: Darlington Kofa

    3m 19s
    Darlington Kofa
  • EP7: max_tx_size

    Authored by: Darlington Kofa

    4m 59s
    Darlington Kofa
0:00
/
~0:00